Abstract
Virtual Screening (VS) methods can considerably aid clinical research, predicting how ligands interact with drug targets. Most VS methods suppose a unique binding site for the target, usually derived from the interpretation of the protein crystal structure. But it has been demonstrated that in many cases, diverse ligands interact with unrelated parts of the target and many VS methods do not take into account this relevant fact.However, this fact increases the computationally complexity exponentially. In this work we enhance the parallelization of non-bonded interactions kernel for VS methods on Nvidia GPU architectures. We show several parallelization strategies that lead to a speed up factor of 15x compared to previous GPU implementations.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
